Alignment Research Engineer Accelerator
We are an ML engineering program that provides the skills, tools, and environment for upskilling in technical AI safety.
ARENA 5.0
We are excited to announce the fifth iteration of ARENA (Alignment Research Engineer Accelerator), a 4-5 week in-person ML bootcamp with a focus on AI safety. ARENA will be running in-person from LISA from 28th April - 30th May. For more information, see here!
Applications to ARENA 5.0 are now closed. To register your interest in forthcoming ARENA iterations, please fill in this form.
Programme
During our in-person bootcamp, participants will work through ML programming exercises together in pairs over 4 -5 weeks under the guidance of expert teaching assistants to develop their skills. Participants will be part of a group of talented engineers and researchers working out of the LISA office in London, allowing them to connect and exchange ideas with active AI safety researchers. In the final week, participants complete a capstone project that dives deeper into a research topic they found compelling during the course.
Curriculum
-
Week 0: Fundamentals
We first cover the fundamentals of deep learning, including PyTorch, CNNs, and ResNets. We’ll dig into the details of how deep learning works through implementing our own optimizers and the backpropagation algorithm.
-
Week 1: Transformers and Mechanistic Interpretability
You will learn how to build and train your own transformers, and learn mechanistic interpretability on transformers, a field which has been advanced by Anthropic’s Transformer Circuits Thread.
-
Week 2: Reinforcement Learning
You will learn the fundamentals of RL, work with OpenAI’s Gym environment to run your own experiments. You will learn about Reinforcement Learning from Human Feedback (RLHF) and apply it to the transformers you built.
-
Week 3: LLM Evaluations
You will learn how to evaluate LLMs. We'll take you through the process of building a multiple choice benchmark from scratch and using this to evaluate current models. We'll then move on to study LM agents: how to build them and how to evaluate them.
-
Week 4: Capstone
We conclude the ARENA programme with capstone projects, where you get to dig into something you found fascinating during the course, applying the skills and knowledge you have accumulated over the last 4 weeks.